Stakeholder Consensus Selection of Data Elements and Indicators for a Pancreatic Cancer Surgical Outcomes Registry in Australia and Aotearoa New Zealand: Modified Delphi Process Authors: Elizabeth Lockie, Anita Skandarajah, John Zalcberg, Kai Brown, Manju Chandrasegaram, Jonathan Koea, Elaine Kuri, David Watters, Andrew Metz, Benjamin P T Loveday, PaCaSO Consensus Group Published: 2026 ABSTRACT ABSTRACTBackgroundThe data elements and indicators in a clinical quality registry (CQR) contribute towards the utility of its output and should be informed by stakeholder input. In Australia and Aotearoa New Zealand (AANZ), there is no CQR for pancreatic cancer surgery. This study aimed to select data elements and registry aims with stakeholder input.MethodsA three‐round modified Delphi consensus process was conducted involving clinicians and consumers. An initial list of data elements was generated through a review of literature, existing international registries, and guidelines. Participants rated the importance of each element using the RAND/UCLA method. In the final round, ≥ 75% consensus was needed for an element to be accepted. Participants rated potential registry aims.ResultsConsumers and clinicians from across AANZ were involved. A total of 66 participants were involved asynchronously across the three rounds (Round 1, n = 33; Round 2, n = 40; Round 3, n = 63), with representatives from consumers, surgery, medical oncology, nutrition, nursing, public health and a consumer organisation. After three rounds, 121 data elements were accepted, and all 9 registry aims were rated with high importance. Of these elements, 54 were classified as quality indicators (structure n = 3, process n = 11, outcome n = 40).ConclusionThe utility of a registry relies on having meaningful and feasible data elements and indicators. Through involvement of key stakeholders in a systematic process, this study has selected data elements and indicators for a proposed AANZ pancreatic cancer surgery CQR. This is an essential step towards establishing such a registry. The next steps will be to pilot the dataset with pancreatic surgery centres across the region, and to establish baseline benchmarks.Keywords: consensus, pancreas, pancreatectomy, pancreatic neoplasms, registry Introduction The Australian Commission on Safety and Quality in Healthcare short‐listed pancreatic surgery as a priority area that would benefit from the development of a national clinical quality registry (CQR) [1], and this is reiterated in the National Pancreatic Cancer Roadmap [2]. There remains a paucity of data about the details of pancreatic surgery and associated treatments around Australia and Aotearoa New Zealand (AANZ) [3]. The lack of data significantly limits the ability of clinicians, consumers, and public health experts to develop local benchmarks and evaluate actual outcomes in this lower volume setting inherent to health care in AANZ. The current pancreatic cancer registries in AANZ [3, 4] do not provide sufficient information regarding pancreatic cancer surgery and do not comprehensively capture data across all or even most institutions performing this type of potentially high morbidity surgery [5]. Therefore, the current registries are unable to deliver the aims of a national CQR. While a small number of AANZ pancreatic surgical units contribute data to the clinically effective American College of Surgeons National Surgical Quality Improvement Program (NSQIP), this system provides benchmarking by comparison across all NSQIP sites (> 90% based in North America), and no public reports exist for an Australian NSQIP collaborative for pancreatic surgery. Thus, there is a need to develop a widely utilised AANZ pancreatic cancer surgical outcomes CQR [5], so that AANZ can reap the clinical, economic and research benefits associated with a CQR. The utility of a CQR relies on having meaningful and feasible data elements and indicators. The Australian Framework for National CQRs (‘the Framework’) emphasises the need to select these with input from key stakeholders, including clinicians and consumers [2]. This should ensure that elements and indicators are relevant, measurable, important for outcomes, and consumer‐centred. Selection of elements and indicators through stakeholder consensus can involve a simple consultation process or a more systematic Delphi consensus process, which involves iterative rounds of questions to reach a consensus. This process has been used by the Monash University‐based Upper Gastrointestinal Cancer Registry (UGICR) [6] as well as other registries [7, 8, 9, 10] to choose elements and indicators reflecting high‐quality care. This study aimed to select data elements and indicators for an AANZ pancreatic cancer surgical outcomes CQR through a modified Delphi process with consumer, clinician and other stakeholder involvement. The process was supported by the AANZ Hepatic, Pancreatic and Biliary Association (AANZHPBA). In addition to selecting elements and indicators, potential CQR aims were rated. Information from this Delphi process will be used to guide data collection at a national level to allow for national benchmarking of pancreatic cancer surgical outcomes. Method Participants A Delphi consensus process was conducted with consumers and clinicians involved in pancreatic cancer surgery from across AANZ, following the CREDES guidelines for conducting and reporting Delphi studies [11]. Participants could partake in any of the rounds without obligation to participate in the others, and participation in at least two rounds meant they could opt to be in the group authorship. Consumers included patients and carers (pre‐ or post‐surgery for pancreatic cancer), who were contacted through three consumer organisations (PanCare, PanKind, Peter MacCallum Cancer Centre consumer group). The consumer organisations used their networks to identify consumers who were interested in the project and provided their contact details with each consumer's permission. The clinicians included pancreatic surgeons contacted through AANZHPBA, with all members of this association being invited, with the final round conducted at the 2025 AANZHPBA Annual Scientific Meeting to facilitate broad participation. Non‐surgeon clinicians were identified for participation through searching the authorship of key research or guidelines about pancreatic cancer care in AANZ, identifying members of organisations involved in pancreatic cancer care (PanCare, PanKind), and through a snowball technique. These stakeholders invited included medical oncologists, radiation oncologists, gastroenterologists, dietitians, nurses, allied health professionals, pathologists and government representatives. Non‐surgeon participants were contacted directly by the investigators via email. Selection of Indicators A list of proposed registry data elements and indicators was prepared through a literature review of studies and guidelines around pancreatic cancer outcomes, guidelines regarding clinical quality registries, and existing pancreatic cancer registries (n = 130). The data elements were categorised into domains to ensure all aspects of pancreatic cancer surgery were covered—pre‐operative, surgical, post operative, pathological, adjuvant treatment, follow‐up and site‐specific indicators. The proposed list was reviewed prior to the Delphi process by a focus group involving four consumers, who accepted all the data elements and added two (‘patient participation in formal prehabilitation programme’ and ‘patient received professional psychological support post operatively’). Delphi Consensus Rounds The modified Delphi consensus process comprised three rounds of voting. Rounds 1 and 2 used online questionnaires, and Round 3 was a meeting conducted in person and online. Responses in all rounds were anonymous. Participants were able to record their name and contact details if they wished to be considered in group authorship, but this was not linked to any voting responses. Rounds 1 and 2 were online questionnaires administered via Qualtrix (Provo, Utah, USA), in which participants were asked to rate the importance of each data element from 1 (not important) to 9 (very important) for the proposed AANZ pancreatic surgery registry. There was an option of ‘unable to answer’ if the participant did not feel they had adequate knowledge or experience to answer. Each element was accompanied by proposed values (e.g., ‘yes’ or ‘no’ for a binary question, ‘millimetres’ for pancreatic duct size) and evidence for the relevance of the element to outcomes in pancreatic surgery. Participants could provide written comments after scoring each element and at the end of the questionnaire. Demographic data were collected in Rounds 1 and 2 regarding participant location (Australian state/territory or Aotearoa New Zealand), age group (18–35, 36–50, 51–65, 66 years or older), gender (male, female, prefer not to say) and role in pancreatic surgery care (clinician/professional, consumer). Clinicians/professionals were additionally asked about their specialty, years in role and hospital setting (public or private). Round 2 included data elements that needed rescoring after Round 1 due to disagreement, borderline importance, or consumer advocacy, and new or significantly changed elements as per participant comments in Round 1. A new section addressed the registry aims (included in response to Round 1 comments), where participants scored 9 proposed aims from 1 (not important) to 9 (very important). The options for registry aims were chosen by the steering committee. The results from Rounds 1 and 2 were analysed using the RAND/UCLA methodology [12]. The median importance (MI) of each data element was calculated, with MI < 7 indicating low importance. The disagreement index indicates the consensus between respondents. It was determined by deducting the interpercentile range adjusted for symmetry (IPRAS) from the interpercentile range (IPR), with an IPRAS > IPR indicating agreement. Items with MI = 7 and disagreement index > 0 were considered ‘borderline’ and were rescored in the next round. Items with any MI and a disagreement index ≤ 0 were rescored in the next round. Borderline elements that were rescored in a subsequent round as MI ≥ 7 and disagreement index > 0 were accepted. Comments from Round 1 were used to clarify wording, divide elements into multiple parts, or add new elements. Round 3 was conducted at the AANZHPBA Annual Scientific Meeting on 17 October 2025 in Perth, Australia. Participants could also partake online instead of in person, and voting from all participants (online and in person) was via the same online platform (Zoom, San Jose, California, USA). It included data elements that needed rescoring from Round 2 and one new element and clarification question suggested from comments. In this round, participants voted (yes or no) as to whether the element should be included in the registry, and an agreement threshold of ≥ 75% was used to establish consensus. Oversight A steering committee provided opinions and suggestions regarding project structure and content. It comprised one consumer and six clinicians from across AANZ, including five pancreatic surgeons and one general surgeon with professional body and government roles in healthcare safety and improvement. The steering committee was chaired by the principal investigator. The final list of data elements was reviewed by the authors, and those used in other registries and publications to benchmark healthcare quality were designated as indicators, and categorised using the Donabedian model as structure, process or outcome indicators [13]. Ethics This study was approved by the University of Melbourne Human Research Ethics Committee (2024‐30023‐57389‐3). Consumers involved in the first focus group received remuneration for their time. Results Round 1 A total of 238 people were invited to complete the Round 1 questionnaire, including 225 pancreatic surgeons via AANZHPBA, 19 other clinicians and 4 consumers. There were 33 responses (13.8% response rate), including 4 consumers and 29 clinicians (25 surgeons, 2 dietitians, 1 nurse, 1 medical oncologist; Table 1). The most common age group was 36–50 years (23/33, 69.7%) and most were male (22/33, 66.7%). The most common location was Victoria (11/33, 33.3%), while each state in Australia and the Australian Capital Territory and Aotearoa New Zealand were represented. For the clinicians, the median years in role was 8 (range: 1–35). The primary hospitals of the surgeons were tertiary or quaternary (24/25, 96.0%) and public (23/25, 92.0%). TABLE 1Demographics of participants.Round 1Round 2Round 3Number334063Location n (%)Victoria11 (33.3)13 (32.5)New South Wales7 (21.2)7 (17.5)Western Australia5 (15.2)5 (12.5)Australian Capital Territory3 (9.1)3 (7.5)South Australia3 (9.1)3 (7.5)Queensland2 (6.1)4 (10.0)Tasmania1 (3.0)2 (5.0)Aotearoa New Zealand1 (3.0)3 (7.5)Age18–351 (3.0)1 (2.5)36–5023 (69.7)29 (72.5)51–655 (15.2)8 (20.0)66 or older4 (12.1)2 (5.0)GenderFemale10 (30.3)9 (22.5)Male22 (66.7)31 (77.5)Prefer not to say1 (3.0)0RoleClinician/professional29 (87.9)38 (95.0)62 (98.4)Consumer4 (12.1)2 (5.0)1 (1.6)Clinician/professionalsRoleSurgeon25 (86.2)33 (86.8)60 (96.8)Allied health2 (6.9)1 (2.6)1 (1.6)Nurse1 (3.4)1 (2.6)0Medical oncologist1 (3.4)1 (2.6)1 (1.6)Healthcare quality and safety01 (2.6)0Consumer organisation01 (2.6)0Years in roleMedian87Range1 to 351 to 45Surgeons—level hospitalTertiary (or quaternary)24 (96.0)31 (93.9)Non‐tertiary1 (4.0)2 (6.1)Surgeons—settingPublic23 (92.0)30 (90.9)Private2 (8.0)3 (9.1) Demographics of participants. From the 132 proposed data elements in Round 1, 84 (63.6%) were accepted (MI > 7 with agreement, some with minor phrasing changes) and 12 (9.1%) were rejected (MI < 7 with agreement or deleted as collapsed into single Charlson Comorbidity Index [14]). The remaining 36 (27.3%) elements had disagreement (n = 11), borderline MI (n = 16), significant changes suggested (n = 8), or low MI with agreement but consumer advocated (n = 1). These data elements moved onto Round 2 to be rescored, in addition to new elements suggested in comments (n = 30). Figure 1 shows the flow of data elements between rounds and Table S1 shows the scoring for each round. Flow chart of data element selection. Round 2 A total of 253 people were invited to complete the Round 2 questionnaire, including 225 pancreatic surgeons via AANZHPBA, 21 other clinicians, and 7 consumers. There were 40 responses (15.8% response rate), including 2 consumers and 38 clinicians (33 surgeons, 1 dietitian, 1 nurse, 1 medical oncologist, 1 representative from public health and 1 representative of a pancreatic cancer consumer organisation; Table 1). The most common age group was 36–50 years (29/40, 72.5%) and most were male (31/40, 77.5%). The most common location was Victoria (13/40, 32.5%), while each state in Australia and the Australian Capital Territory and Aotearoa New Zealand were represented. For the clinicians, the median years in role was 7 (range: 1–45). The primary hospitals of the surgeons were tertiary or quaternary (31/33, 93.9%) and public (30/33, 90.9%). In Round 2, 66 data elements were scored (Figure 1, Table S1). Of these, 37 (56.1%) were accepted (MI > 7 with agreement) and 9 (13.4%) were rejected (MI < 7 with agreement). An additional two elements were rejected as they were the less favoured option in a pair of similar elements that participants were asked to score preferentially (‘pancreatic duct size based on pre‐operative imaging’ less favoured to ‘pancreatic duct size based on intra‐operative measurement’, and ‘intensive care unit admission >24 hours’ less favoured to ‘unplanned readmission to intensive care unit’). Five of the data elements with borderline importance also had MI = 7 in Round 1, so were accepted. There were 18 elements that needed to proceed to Round 3 for rescoring, including 13 with disagreement and 5 with borderline importance. The section in Round 2 regarding registry aims was added due to comments in Round 1 that the choice of data elements would depend on the aims of the registry. Of the nine proposed aims, each was scored with MI > 7 with agreement (Table 2). TABLE 2Registry aims.AimMedian importanceDisagreementMonitor short‐term surgical outcomes9.0NMonitor long‐term surgical outcomes9.0NMonitor short‐term oncologic/cancer outcomes9.0NMonitor medium‐to‐long‐term oncologic/cancer outcomes9.0NBi/national benchmarking9.0NAssess adherence to guidelines and optimal care pathways7.0NSupport research9.0NFacilitate data sharing and collaboration between centres9.0NProvide a framework for clinical trials8.5N Registry aims. Round 3 Round 3 included 63 participants in a hybrid setting of in person meeting and online, and included 60 surgeons, 1 medical oncologist, 1 dietitian and 1 consumer (Table 1). Other demographic data were not collected in this round. Round 3 required voting on 19 data elements, including 18 that required rescoring from Round 2 and one new element suggested in Round 2 comments (‘pancreatic gland thickness’). Two elements were accepted as they had a consensus of ≥ 75%: ‘pre‐operative haemoglobin (closest to surgery)’ (45/60, 75%) and ‘peak C‐reactive protein in first 7 days post‐surgery’ (48/62, 77%). The remaining 17 (89.5%) were rejected (Figure 1, Table S1) due to < 75% agreement. Participants were also asked, ‘should frailty and functional health status apply to all ages (rather than ≥ 65 years only)’? as there had been multiple comments that these measures should apply to all. The agreement threshold was not reached (45/63, 71.4%), meaning these measures would only be applied to those aged ≥ 65 years of age. At the end of the Delphi process, 121 data elements had been accepted (Table 3), and a total of 66 participants (6 consumers, 60 clinicians) had been involved. Data elements used for benchmarking in existing pancreatic surgery CQRs in other countries were identified from the final list, resulting in 54 being classified as quality indicators (structure n = 3, process n = 11, outcome n = 40; Table S2). The remaining 67 data elements related to demographics, risk adjustment or site characterisation. TABLE 3Final data element list.Pre‐operative data elementsHeight and weight at admissionAmerican Society Anaesthesiologists (ASA) scoreBleeding disorders including use of anticoagulation or antiplatelet medicationCa 19.9 at diagnosisCa 19.9 post neoadjuvant treatment (closest to surgery)Charlson Comorbidity Index (CCI)Date of diagnosisEastern Cooperative Oncology Group (ECOG) scoreHistory of IPMNIf ≥ 65 years: Functional health statusIf ≥ 65 years: Rockwood clinical frailty scaleImmunosuppressive therapyNeoadjuvant chemotherapyNeoadjuvant radiotherapyParticipation in formal prehabilitation programmePost neoadjuvant therapy resectability statusPre‐operative dietitian review and/or nutritional supportPre‐operative albumin (closest to surgery)Pre‐operative biliary stentPre‐operative bilirubin (closest to surgery)Pre‐operative endoscopic ultrasoundPre‐operative haemoglobin (closest to surgery)Pre‐operative INR (closest to surgery)Pre‐operative resectability status (before neoadjuvant treatment)Pre‐operative tissue diagnosisPre‐treatment discussion at multidisciplinary meetingResponse to neoadjuvant therapyWeight loss in the 3 months pre surgerySurgical data elementsHospitalDate of surgeryCase acuityPlanned surgical intentType of surgery performedOperative approachPancreatic gland textureTechnique used to divide pancreasUse of adjunct around pancreatic anastomosis to reduce risk of post operative pancreatic fistulaPancreatic reconstruction techniqueIntra‐operative placement of pancreatic duct stentIntra‐abdominal drain(s) placedArterial resectionVenous resectionExtra‐pancreatic organ resectionOperation durationIntra‐operative blood lossIntra‐operative blood transfusionIntra‐operative complications based on Class Intra grading systemSynoptic operative report usedType of antibiotic used for surgical prophylaxisRegional anaesthesia (spinal or epidural)Braun enterostomy usedPancreatic duct size based on intra‐operative measurementIf stapler used to divide pancreas—detailsPost operative data elementsPeak drain fluid amylase post operative Day 1 and Day 3 and Day 5Day of final drain removalDay of nasogastric/nasojejunal tube removalDay of resumption of solid dietIntubation > 48 hSuperficial incisional infectionIntra‐abdominal infectionSurgical wound dehiscenceHospital acquired pneumoniaNew deep vein thrombosis (DVT) and/or pulmonary embolism (PE) requiring therapeutic anticoagulationOrgan failurePost operative acute kidney injuryPost operative stroke (cerebrovascular accident or transient ischaemic attack)Post operative myocardial infarctionPost operative sepsisPost operative deliriumPost operative pancreatic fistula (clinically relevant)Post pancreatectomy haemorrhagePost operative delayed gastric emptyingPost operative pancreatitisPost operative radiologic interventionPost operative endoscopic interventionReturn to operating theatre within 30 daysClavien Dindo complication ≥ 3Comprehensive complication index at 30 days post‐surgeryIn hospital mortalityWeight at dischargeDate of dischargeDischarge destinationEnd of life/withdrawal of care within 30 days post opReadmission within 30 and 90 days (excluding admissions for adjuvant therapy)Enhanced Recovery After Surgery (ERAS) protocol used for individual patientIf ≥ 65 years: discharge home with new services at homeIf ≥ 65 years: functional health status on dischargeUnplanned readmission to ICUPeak C‐reactive protein in first 7 days post‐surgeryPathological data elementsHistopathological diagnosisPathology locationDisease stageTumour sizeLymph node count and number of positive lymph nodesSynoptic pathology report usedMargin statusNumber of margins reportedSpecific margins reportedClosest margin (mm)Perineural invasionLymphovascular invasionMultiple tumour fociTumour differentiationAdjuvant treatment data elementsAdjuvant chemotherapyAdjuvant radiation therapyFollow‐up data elementsDate of last reviewStatus at last reviewDate of deathRecurrencePatient reported outcome measure completedDevelopment of post pancreatectomy diabetesIf ≥ 65 years: Frailty at 1 yearSite‐specific data elementsNumber of pancreatic surgeonsNumber of pancreatic resections per yearMultidisciplinary team membersFrequency of on‐site multidisciplinary meetingSite setting (public or private)Enhanced recovery after surgery (ERAS) program for pancreatic resection patientsOn‐site servicesPrehabilitation service available for pancreatic resection patientsAvailability of local follow‐up for rural or regional patients who have surgery at this site Final data element list. Discussion This Delphi consensus process utilised three rounds of questions with consumer and clinician stakeholders to select 121 data elements, of which 54 may be used as indicators to measure pancreatic cancer surgical quality in AANZ. The data elements will form the basis for the proposed CQR. Registry aims were also rated by participants, with wide support to measure and benchmark short‐term surgical and oncologic outcomes, and to facilitate research. This is the largest consensus study in AANZ to determine how to measure pancreatic surgery structure, process, and outcomes, and is essential in the development of an AANZ pancreatic cancer surgical CQR. An important step in optimising a CQR is having a valid set of indicators. Indicators should be mapped back to a domain of care to determine where all aspects of care are covered by the data collection, as simply having a ‘set of valid indicators does not guarantee a valid set of indicators’ [15]. By involving consumers and a breadth of clinicians, this project has developed a list of indicators that span each domain of pancreatic cancer surgery, from pre‐operative to follow‐up indicators, as well as site‐specific indicators. A key strength of this study was the breadth of stakeholders involved in the process. The Framework emphasises the importance of registry development and data element and indicator selection in consultation with key stakeholders [2]. Lack of consumer involvement was highlighted by Maharaj et al. as a limitation in the selection of indicators for the UGICR [6]. In terms of geography, there was representation across all states of Australia, the Australian Capital Territory, and Aotearoa New Zealand. This Delphi process incorporated opinions from clinicians with a range of experience, and who were based in differing hospital contexts. This broad approach was used to avoid overrepresentation of opinion from high‐volume surgeons based in better‐resourced academic centres. Multiple participants in Round 1 commented that their opinions regarding the importance of data elements would depend on the registry aims. This reflects the multiple purposes a CQR can have, such as monitoring short or long‐term surgical and oncological outcomes or providing a research platform. It was noteworthy that all the proposed registry aims, which were broad, were rated as important with agreement. Stakeholders want a CQR for AANZ that has broad applications, meaning the indicators in the registry need to be comprehensive. This Delphi Consensus selected 121 data elements, which is in the middle of the range for European pancreatic surgery registries (16–285) [16], and fewer than the 314 for the combined base and pancreatectomy modules of NSQIP [17, 18]. This suggests the number selected is a feasible number of data elements to collect. Planning for a national CQR must rigorously outline data collection compliance and maintenance, with direction from the national Framework for CQRs [2]. There are certain limitations to this study. While non‐surgeon clinicians participated, some clinician craft groups who were invited did not respond, including radiation oncologists, pathologists, radiologists and registry experts. Additionally, no medical oncologists responded. However, the high proportion of surgeons may not be a significant limitation given this registry is focussed on surgery‐specific care. Existing pancreatic cancer registries in Australia have a broader oncology focus, and the data elements selected in this study may be complementary to those already collected. Conclusions A robust binational pancreatic surgery CQR is an essential step forward to optimise pancreatic cancer surgery in AANZ. Careful selection of registry data elements and indicators that are consumer‐centred, useful and feasible will optimise the output such a registry will bring. This Delphi Consensus study employed a selection process involving key stakeholders to determine a list of 121 data elements, of which 54 can be used as quality indicators, and elucidated the potential registry aims. The next steps towards a binational AANZ pancreatic cancer surgical CQR include collection of retrospective data based on these data elements to derive baseline benchmarks and to establish sustainable solutions for an ongoing CQR. Funding This project was funded in part by the Epworth Medical Foundation. Elizabeth Lockie is the recipient of a University of Melbourne Faculty of Medicine, Dentistry and Health Sciences Graduate Research Trust Scholarship. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Table S1: Data element scoring. Table S2: Data dictionary. Acknowledgements The authors would like to thank all the participants in the project for their invaluable input. We are grateful for the assistance from Pancare, PanKind and Peter MacCallum Cancer Centre for identifying consumers to participate. This project was endorsed by and undertaken in conjunction with the Australian and Aotearoa New Zealand Hepatic Biliary and Pancreatic Association (AANZHPBA). Open access publishing facilitated by The University of Melbourne, as part of the Wiley ‐ The University of Melbourne agreement via the Council of Australasian University Librarians. This project was funded in part by the Epworth Medical Foundation. Elizabeth Lockie is the recipient of a University of Melbourne Faculty of Medicine, Dentistry and Health Sciences Graduate Research Trust Scholarship. The PaCaSO group authorship includes: Grahame Gare, Alistair Rowcroft, Ankit Jain, Arul Suthananthan, Bernadette Goodwin, Brett Knowles, Claire Stevens, David Cavallucci, Emma Downie, Geoffrey Yuet Mun Wong, Izhar ul Haque, Jamie Crichton, Joel Lewin, Kai Brown, Kate Furness, Kaye Bowers, Marwan Idrees, Neil Merrett, Nick Bull, Nick O'Rourke, Russell Hodgson, Savio George Barreto, Jane‐Louise Sinclair, Matilda Anderson, Soon Lau, Di Chambers, Anthony Farfus, Christopher Nahm, Christopher Seng Hong Lim and Laurence Webber. Contributor Information Elizabeth Lockie, Email: elockie18@gmail.com. PaCaSO Consensus Group: Grahame Gare, Alistair Rowcroft, Ankit Jain, Arul Suthananthan, Bernadette Goodwin, Brett Knowles, Claire Stevens, David Cavallucci, Emma Downie, Geoffrey Yuet Mun Wong, Izhar ul Haque, Jamie Crichton, Joel Lewin, Kai Brown, Kate Furness, Kaye Bowers, Marwan Idrees, Neil Merrett, Nick Bull, Nick O'Rourke, Russell Hodgson, Savio George Barreto, Jane‐Louise Sinclair, Matilda Anderson, Soon Lau, Di Chambers, Anthony Farfus, Christopher Nahm, Christopher Seng Hong Lim, and Laurence Webber Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Supplementary Materials Table S1: Data element scoring. Table S2: Data dictionary. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.